diff --git a/build.py b/build.py
index 91ba43d38d..76d86e3b52 100755
--- a/build.py
+++ b/build.py
@@ -962,7 +962,6 @@ def create_dockerfile_buildbase_rhel(ddir, dockerfile_name, argmap):
             libcurl-devel \\
             libb64-devel \\
             gperftools-devel \\
-            patchelf \\
             python3-pip \\
             python3-setuptools \\
             rapidjson-devel \\
@@ -990,7 +989,8 @@ def create_dockerfile_buildbase_rhel(ddir, dockerfile_name, argmap):
           wheel \\
           setuptools \\
           docker \\
-          virtualenv
+          virtualenv \\
+          patchelf==0.17.2
 
 # Install boost version >= 1.78 for boost::span
 # Current libboost-dev apt packages are < 1.78, so install from tar.gz
@@ -1089,7 +1089,6 @@ def create_dockerfile_buildbase(ddir, dockerfile_name, argmap):
             libcurl4-openssl-dev \\
             libb64-dev \\
             libgoogle-perftools-dev \\
-            patchelf \\
             python3-dev \\
             python3-pip \\
             python3-wheel \\
@@ -1110,7 +1109,8 @@ def create_dockerfile_buildbase(ddir, dockerfile_name, argmap):
 RUN pip3 install --upgrade \\
           build \\
           docker \\
-          virtualenv
+          virtualenv \\
+          patchelf==0.17.2
 
 # Install boost version >= 1.78 for boost::span
 # Current libboost-dev apt packages are < 1.78, so install from tar.gz
@@ -1354,11 +1354,13 @@ def dockerfile_prepare_container_linux(argmap, backends, enable_gpu, target_mach
         libcurl-devel \\
         libb64-devel \\
         gperftools-devel \\
-        patchelf \\
         wget \\
         python3-pip \\
         numactl-devel
 
+RUN pip3 install --upgrade pip \\
+    && pip3 install patchelf==0.17.2
+
 """
     else:
         df += """
@@ -1562,7 +1564,9 @@ def add_cpu_libs_to_linux_dockerfile(backends, target_machine):
 
 # patchelf is needed to add deps of libcublasLt.so.12 to libtorch_cuda.so
 RUN apt-get update \\
-      && apt-get install -y --no-install-recommends openmpi-bin patchelf
+      && apt-get install -y --no-install-recommends openmpi-bin
+RUN pip3 install --upgrade pip \\
+    && pip3 install patchelf==0.17.2
 
 ENV LD_LIBRARY_PATH /usr/local/cuda/targets/{cuda_arch}-linux/lib:/usr/local/cuda/lib64/stubs:${{LD_LIBRARY_PATH}}
 """.format(
diff --git a/docs/user_guide/jetson.md b/docs/user_guide/jetson.md
index e2b2b0ad34..7ecab6a2b2 100644
--- a/docs/user_guide/jetson.md
+++ b/docs/user_guide/jetson.md
@@ -1,5 +1,5 @@
 <!--
-# Copyright 2021-2024, NVIDIA CORPORATION & AFFILIATES. All rights reserved.
+# Copyright 2021-2025, NVIDIA CORPORATION & AFFILIATES. All rights reserved.
 #
 # Redistribution and use in source and binary forms, with or without
 # modification, are permitted provided that the following conditions
@@ -82,7 +82,6 @@ apt-get update && \
             libtool \
             libboost-dev \
             rapidjson-dev \
-            patchelf \
             pkg-config \
             libopenblas-dev \
             libarchive-dev \
@@ -95,7 +94,7 @@ apt-get update && \
 Additional Onnx Runtime dependencies must be installed to build the Onnx Runtime backend:
 
 ```
-pip3 install --upgrade flake8 flatbuffers
+pip3 install --upgrade flake8 flatbuffers patchelf==0.17.2
 ```
 
 Additional PyTorch dependencies must be installed to build (and run) the PyTorch backend: